Na'vi Rosetta Stone possible with some help

[Ezy Ryder]

If you need some music to the background, I could help. I'm also a CPP programmer.

[Nume fpi sänume]

Awesome, good to know there is still interest in this. I didnt want to drop it, so seeing that there are still people with an idea for it, but were just waiting is good enough for me

[Ezy Ryder]

I could write a CPP version of this. Just somebody would have to make:
-A complete lesson plan
-A complete verbs used in lesson
-A complete vocal part (as long as You want vocal in it)
-A complete images part
I could take the rest (programming).

[Nume fpi sänume]

You should contact Irongamer and collaborate with him, hes got the Na'vi Language Tool up, which could speed things up, or at least give you something to work with.

[Ezy Ryder]

Besides I need to know how would You want the app to look like. Than (if it's virtually possible) I'll do my best.

Ps.: I've written the core already (first 350 lines of code).

[Eyaye Tskxe]

So I started with graphics today with "Pxuntil", and realized varied we could make this. I want to keep it simple, with just one color backgrounds, but was wondering if anyone had input on the type of background. I came up with five that I think look alright. Our choice, however, depends on what the actual program will look like; specifically, it's background colour. Let's assume it's black for now. Background refers to the area right around the object, box is the box around the object.

(Note: I have no idea how Rosetta Stone looks, I have never used anything like it before. This is merely a mock up to see how these might look. Learn Na'vi text borrowed without permission from learnnavi.org)

Option 1: White background, White box. I personally don't like it, but I thought I'd throw it up there.


Option 2: Object has 100% Opacity "Outside-the-lines" background, transparent box.


Option 3: Object has 57% Opacity "Outside-the-lines" background, transparent box.


Option 4: Object has a 57% Opacity box with a circular feather.


Option 5: Object has a 100% Opacity box with a circular feather.


It is difficult to see the effect the transparency had on a black background, but I assure you, it is there. Transparencies can be easily raised or lowered, 57% was merely what I set them at here.

I prefer option 2. It's easy to do, and looks pretty nice for being intentionally messy. Any other opinions? Thoughts? Criticisms?

[Ezy Ryder]

I prefer option 2 or 5. To make the first demo I'll need the logo (the Learn Na'vi thing at the top), drawings of the words visualizations and the exact places of the squares (text one and image one). Then I'll upload the demo in here.

Ps.: You want it to be a full-screen app or windowed. If windowed I've got to know the exact size.

[Eyaye Tskxe]

I'll PM SeaBass about getting formal permission to use the logo. I haven't had time to crank out more pictures, but I'll see if I can't get a template up. It probably won't be until tomorrow night, though.

Let's start off with just a windowed app. I'm thinking 700 (W) by 600 (H) pixels should be good, at least for the demo.

[Ezy Ryder]

I've got just one more question, before I'll end the writing the demo. What should it do? I mean, only show the picture and text in English/Na'vi? I didn't ever used Rosetta Stone so I don't know too much 'bout it.

[Miyapxi]

I'll PM SeaBass about getting formal permission to use the logo. I haven't had time to crank out more pictures, but I'll see if I can't get a template up. It probably won't be until tomorrow night, though.

Let's start off with just a windowed app. I'm thinking 700 (W) by 600 (H) pixels should be good, at least for the demo.

From what I've seen the entire process is done in the language being learned.  By using words and pictures rather than words and words you learn to think in the language and recognize by image rather than translate quickly in your head.

[Eyaye Tskxe]

I, as well, am not sure on how this is laid out. Should we, for starters, merely have two boxes that show the word and a corresponding picture? Maybe have the picture or the word be blanked out until someone hits a button to go forward?

In addition, I haven't heard back from Seabass about the logo yet, so we won't have that right away. I won't have the GUI or more pictures until later tonight, I've been a bit busier than I thought. Though I should have expected it... it's been finals week. :[

Well, 2 in the morning is close to "night", I guess. I wasn't really sure what to put on this; so I went super simple. All it's got are two boxes for words, and a button to play the audio clip, as well as a nice looking border. It ended up being 800 X 600, I saw that that was a default size and went with it.

(Learn Na'vi logo still used without permission; I haven't yet received a message condoning or condemning its usage.)

Spoiler

I'm thinking once we're further along, we could put buttons along the bottom bar for easy access. Most likely things like "Menu" and "Lessons", etc.

I've yet to draw any more pictures; the search feature on Adobe Reader broke temporarily, crippling my ability to easily find more words to use. Hopefully by Monday I'll have a good list going. (Also, if anyone has the time and wants to contribute sketches or even ideas on how to show more abstract terms like "See", that have more-than-physical meanings, feel free to throw them out there. Many hands make light work!)

[JayWalker]

Is this project still underway? I was talking with Tiriuä in TS the other day about making a Na'vi Rosetta stone type program, and we were unsure if this one is still making progress. I was thinking about writing an app like this in VB6 and have the lessons basically be modular, so additional lessons/images/sounds/etc. could be added on at any time. I could probably have the app and a starter lesson done within a week or two. Just curious if I should bother with this myself or if this one is still going well.

[Eyaye Tskxe]

I haven't been able to draw much anymore. School and sports got in the way big time. So there's been pretty much no progress on the pictures side of things. I don't know about anything else, such as programming.

[Mithcoriel]

So, lemme summarize what a Rosetta stone program actually does: (are there variations of this?)
4 Images appear. E.g. a man, a woman, a cat, and a dog
One of those words is written on top and spoken, in the language to be learned. E.g. "cat". If the user clicks on any image but the cat, a cross appears. If he clicks on the cat, a tick appears that he got it right. Then the next word appears, once again automatically spoken by the computer, till all words have been gone through. Then the next set of 4 words appear.

Images - It would be fun and logical to use images from the movie, but not very wise considering copyright issues.

Would it really? I mean, we're already using screen grabs for the avatars in the forums here. How is that any different? Just sorta wondering..

For starters, the pronouns are a must to learn. So begin with the pronouns. I doubt we will need pictures for this, as it's only five words, and easy to memorize.
Second, simple verbs. "To be", "Give", etc.
After that, markers of intransitive verbs, since those help form basic sentence. No need to go into any verb tenses other than present this early in the game.
Next, I'd go with showing possession; again, it's helpful in simple sentences. ("My mate is hawt.")
And then Datives, or showing the recipient of something ("She gave Austin the rock")

But, that's not really how a rosetta stone program works. Unless you've been using some version I'm not familiar with. Like I said, the program I know focuses mostly on vocab. The first few groups of four pictures give you only vocabulary. Then they start forming simple sentences, e.g. once you've learned the word for "cat" and "car", there's one called "cat and car", showing both. (Which makes you learn the word "and" as well as repeating those two words).  Then, once you've learned the word for "boy" and "table", there's a set with "boy is on table", "boy is under table", "boy is next to table" etc. etc.

Let me throw around some ideas of which words should come first, together in groups of fours. Might be a good idea to start thinking of that, for those of us not otherwise involved in the project. Feel free to correct and improve this first very rough idea.

- Na'vi, Ikran, Tree, (and, what else? Sky?)
- Male person (Tutan), Female person (Tutee), maybe Sky Person (human. Easy to derive if the word "sky" was previously learned. But not absolutely necessary), and possibly a pic with two things in it, like "Na'vi and Ikran".
- At some point, maybe here, maybe later, a set of four pandoran animals. Like Palulukan, Angsit (or what was that hammerhead called?), Toruk, Pa'li/Horse
- At some point, a group of natural things, like "stone", "fire", "wind" and whatever we have that fits in here.
- Preposition, maybe. "Na'vi is in front of/beside/in/on tree". Just pick four prepositions we know the Na'vi meaning of
- Parts of the face. Eye, nose, ear, lips (since we have a word for "lips" but not for "mouth")
- In the original Rosetta stone, they teach the numbers by showing one image with the numbers "1,2,3", the next one with the numbers "1,2,3,4", the next one with the numbers "1,2,3,4,5" and so on. Seems weird, cause even someone who doesn't know the numbers could just count how many numbers were in that sentence, but it seems to work. They worked this out, they must have a good reason for it.
- They need to learn the plural, how it's got that "ay", and the softening of the consonants.  E.g. one pic titled "Sky person", one pic titled "four skypeople", one pic titled "Na'vi", another titled "Five Na'vi". Repeat this for the next couple of exercises, with various words, to give the person an intuitive feel for the plurals, while also repeating the numbers.
- Practice Dual and Trial. One image shows the eye (nari), as exercised previously. Another (menari, the dual form) shows two eyes belonging to the same person. Another shows two people's faces side by side, so that one eye of each is visible, and this is titled "Two eyes", but it's not the dual form. And then maybe one showing a creature with three eyes, for the trial, and one showing lots of eyes, for the plural.
- The rosetta stone program then has words such as "fly" and "swim" (don't think we have the latter in Na'vi yet. We could replace it with "walk"). Well, basically a bunch of movement verbs. We could have: "The Ikran flies", "The Ikran walks", "The Na'vi rides the Ikran", and, dunno, I'm sure we'll find a fourth one to fill in here. "Dance", for example.
- The program then has some simple sentences, along the lines of "is the car red? Yes, it's red", "Is the car blue? No, it's yellow". So we'd just have to use some vocabulary that was previously introduced, to make these sentences. What's good here is they'd automatically learn the word "srak" (at the end of yes-or-no questions) in the process. Could be "Is the Na'vi walking? No, he's riding" or whatever.
- Words like hunt, kill, and die could fit together too. For "the Na'vi hunts the Yerik", you could have a Na'vi sneaking up on one, then "The Na'vi kills the Yerik" with him clearly slicing the throat with the dagger, and maybe one "The Na'vi dies" or "The Yerik dies" where the creature perishes in some other way, like falling down a cliff, to show it's not the same as killing.
- We've got words for mother, father, daughter, son, so might as well add those together. Make some Na'vi stand in a "family portrait" position, and each pic focuses on one of them, maybe 4 times the same pic with a different member encircled.
- It's true, the words "give" and "be" are useful. We could have a picture titled "Jake is a human", showing Jake in his human form, and one going "Jake is a Na'vi", in his Na'vi form, and just to keep the memory of the word "and" fresh, one that shows two people, saying "X and Y are Na'vi".
- And the words "give" and "take" can be exercised in the next group, with various "X gives Y the object Z", and "X takes object Y (possibly FROM person Z, so we can learn the "from")". Perhaps, to make the Dative endings clearer, have two pictures that are reversed, e.g.  one that goes "X gives something to Y" and then "Y gives something to X".
- Future and past forms of verbs are slightly difficult to show visually. We could have a Na'vi just getting onto his Ikran for "Na'vi will fly", just gettig off it for "Na'vi has flown". But those are the immediate past and future. For the perfective and imperfective, we could use "eat", where a person is in the process of eating in one pic, and in the other, he's just finished (obviously, for that purpose the word "eat" should be added in one of the previous exercises)

Well, just sort of a thought. More ideas?

[Eyaye Tskxe]

That is a lot more comprehensive then anything I had thought, irayo! I've never actually used Rosetta Stone, I just knew that it used pictures, so your post was very enlightening. But I don't really know how to respond to it... I don't know programming at all, so I don't know how easy/difficult it will be to implement.

[Mithcoriel]

I think programming it should be fairly easy. At least it's a simple algorithm, compared to some other crazy things one could program. But I never programmed in flash, so I don't know for sure..¨

But anyway, you don't need to know anything about programming to join the discussion on what the content should be. That's sort of a pedagogic question ^^.

[JayWalker]

Hey guys. Forgive me for suggesting a hijacking of the progress made over here.... but check out the "Tree Of Voices" thread I made here in the projects section. It is essentially what this project was intended to be, but I've released a useable version of the program. The reason I post here is because I saw that lots of people were gathering images and whatnot for use in this program. These gathered images which are currently not doing much, can be VERY EASILY made into lessons to be used with my program. I hope some of you choose to take part in this so that the mutual want of a rosetta-stone like program can make further progress

[Hyperion]

ooh ooh sign me up. just started wont be any help on development and dont have any versions of rosetta stone. how would i be able to get in on getting help in this way

[Ezy Ryder]

I'm programming in C++ and I'm already working on a demo. I could make a connections .h library and a simple ask and check algorithm in a simple loop. As I said it will be written in C++ which means that it will be in .exe format. You'll be able to use it on any Microsoft's platform such as Windows and even phones and palmtops (as long as it is working on a Microsoft's platform). I can write the program in 1 hour, implement pictures, sounds and words in (depends on how many will we have) 1 hour and make all alfa-tests in 30 minutes. Then I'll need some beta-testers and after beta-tests I'll put the most actualized version in this topic. Right now I'll need full word-bank or minimally know how many words do You want the program to teach.

Ps.: I could also make a Polish version.
Ps2.:I think that as long as we don't make money out of it, it's legal. Na'vi language is Fox's property if I'm right.

[Mithcoriel]

Cool. Are you gonna make the code open-source?

Well I'll beta-test if you still need someone.

Can you program it in a way so that new lessons can easily be added? Sort of the way JayWalker did, where you add a new folder into the "Lessons" folder, make a little textfile with specifications, and add the files needed? Cause that would be pretty useful.

I think that as long as we don't make money out of it, it's legal. Na'vi language is Fox's property if I'm right.

I don't think the Na'vi language is the issue, but the pictures, for example. If we just get em from google, they might be copyrighted.